« The Link between Writing Style and Elite Consensus: Obergefell v. Brown | Main | Saving Greece via Crowdfunding? »

Tuesday, June 30, 2015

Strange Bedfellows #12: Closing Thoughts on The Science of Learning

This post is part of the Strange Bedfellows series.

In this guest blogging series, I’ve had fun identifying connections between canonical cases not ordinarily taught together.  But is it any more than a parlor game? 

For a teacher with years of familiarity with the cases, it can be invigorating to rearrange the furniture—and students always benefit from an invigorated teacher.  The fear is that giddily wandering away from the orthodox doctrinal silos might create an obstacle for students who need to know and apply the orthodox solutions to problems.  I’m convinced that when handled properly, exposure to these strange bedfellows actually leads to better comprehension of the orthodox approach.  This is primarily because a semester with a healthy amount of unexpected (but valid) juxtapositions will inevitably take advantage of two strategies favored by researchers into the science of learning: interleaving distinct but related topics, and repeated exposures spread over time.

The silo approach (a unit on the Commerce Clause, followed by a unit on the Spending Clause, and so on) presumes that it would be unduly confusing for students to shift gears, hurting their comprehension. But studies show the opposite: interleaving the presentation of related but distinct topics results in better mastery of each topic. Learners understand the relationships among silos better, and also—perhaps unexpectedly—they understand each silo better.

In Make It Stick: The Science of Successful Learning (2014), authors Peter Brown, Henry Roedinger, and Mark McDaniel describe a study that compared different methods to teach students how to identify works by different painters:

Researchers initially predicted that massed practice in identifying painters’ works (that is, studying many examples of one painter’s works before moving on to study many examples of another’s works) would best help students learn the defining characteristics of each artist’s style. Massed practice of each artist’s works, one artist at a time, would better enable students to match artworks to artists later, compared to interleaved exposure to the works of different artists. The idea was that interleaving would be too hard and confusing; students would never be able to sort out the relevant dimensions. The researchers were wrong. The commonalities among one painter’s works that the students learned through massed practice proved less useful than the differences between the works of multiple painters that the students learned through interleaving. Interleaving enabled better discrimination and produced better scores on a later test that required matching the works with their painters. The interleaving group was also better able to match painters’ names correctly to new examples of their work that the group had never viewed during the learning phase.

Similar results occurred in a study teaching people how to identify different families of birds, how to hit different kinds of pitches, and how to solve different kinds of math problems. These tasks strike me as similar to what we expect law students to do: transfer the knowledge gained through study of past cases to help identify, categorize, and resolve issues when they arise in previously unseen circumstances.

However, the studies showing the power of interleaving also reveal a cognitive illusion: students who learn interleaved material routinely underestimate their progress when compared to the silo method. This is largely because the advantages of interleaving tend to reveal themselves slightly later in time.  From Make It Stick:

The learning from interleaved practice feels slower than learning from massed practice. Teachers and students sense the difference. They can see that their grasp of each element is coming more slowly, and the compensating long-term advantage is not apparent to them. As a result, interleaving is unpopular and seldom used. Teachers dislike it because it feels sluggish. Students find it confusing: they’re just starting to get a handle on new material and don’t feel on top of it yet when they are forced to switch. But the research shows unequivocally that mastery and long-term retention and are much better if you interleave practice than if you mass it.

In my experience, students actually do not dislike the type of interleaving described in these blog posts and in my casebook, so long I am transparent with them about the logic. A few months into the semester they can feel the benefits of better comprehension and retention as they solve problems across silos. By the end of a semester, they know they are further ahead than they would have been, despite the initial feeling of unfamiliarity.

One advantage of interleaving is that it forces some delays and spreads given material over a larger stretch of time.  Instead of studying everything about the Commerce Clause in a one week silo, it is studied a bit at a time over several weeks or months. Repeated work with a topic over time, with enough lapse between exposures for a little forgetting to occur, improves a student’s ability to later retrieve and apply the knowledge.  This passage from Make It Stick describes some of the research:

For a vivid example [of the benefits of spacing out lessons] consider this study of thirty-eight surgical residents. They took a series of four short lessons in microsurgery: how to reattach tiny vessels. Each lesson included some instruction followed by some practice. Half the docs completed all four lessons in a single day. … The others completed the same four lessons but with a week’s interval between them.

In a test given a month after the last lesson, those whose lessons had been spaced a week apart outperformed their colleagues in all areas—elapsed time to complete a surgery, number of hand movements, and success at reattaching the severed, pulsating aortas of live rats. The difference in performance between the two groups was impressive. The residents who had taken all four sessions in a single day not only scored lower on all measures, but 16 percent of them damaged the rats’ vessels beyond repair and were unable complete their surgeries.

Why is spaced practice more effective than massed practice? It appears that embedding new learning in long-term memory requires a process of consolidation, in which memory traces (the brain’s representations of the new learning) are strengthened, given meaning, and connected to prior knowledge—a process that unfolds over hours and may take several days. Rapid-fire practice leans on short-term memory. Durable learning, however, requires time for mental rehearsal and the other processes of consolidation. Hence, spaced practice works better. The increased effort required to retrieve the learning after a little forgetting has the effect of retriggering consolidation, further strengthening memory.

In addition to Make It Stick, interested readers can consult two free online books about the science of learning: 

Posted by Aaron Caplan on June 30, 2015 at 11:38 AM in Teaching Law, Things You Oughta Know if You Teach X | Permalink

Comments

I wonder at what level interleaving is optimal in a particular course. For example, in a Contracts course, would it help to study a case on consideration and one on offer and acceptance in the same class, because they often are related (though distinct)? Or is it enough that a single class addresses two distinct kinds of consideration problems, or even addresses the same issue but encounters two cases that draw different conclusions and thus provides an opportunity for synthesizing the cases and identifying the differences that explain the differing outcomes? Interleaving, at least as it is described in the essay, should not lead to the conclusion that one should mix consideration with illegal contracts simply because both are "related" to contract law. So, the level at which one interleaves is a crucial part of the pedagogy. Indeed, comparing the styles of different painters in the same class session might be roughly analogous to comparing the opinions of two courts, even though they address precisely the same issue, not to mention the same topic.

Posted by: Charles Calleros | Jul 1, 2015 5:20:48 AM

You make a good observation that there is already some interleaving of topics within the traditional model of law teaching. Finding the optimal amount will require trial and error, but I suspect that compared to most people's current teaching, more interleaving would move us closer to the optimum. Good opportunities for interleaving arise when teaching from fact patterns where topics are litigated together (just like the kinds of fact patterns you develop for final exams). You mentioned consideration being logically connected to offer/acceptance, so look for cases where both issues arise. To work through such a case will require discussion of both doctrines to reach the ultimate result, making the interleaving seem entirely fluid and natural. With the right cases to assign, you can move from this format (week 1 is consideration, week 2 is offer/acceptance) to this format (two weeks on the broader topic of "forming a contract" that will acquaint students with two doctrines--but in a way that allows them repeated exposure to each doctrine spread over more time).

Posted by: Aaron Caplan | Jul 2, 2015 12:31:51 PM

Hi Aaron: I applaud your courage for teaching a course and writing a casebook using an interleaving model. I took a seminar last year that discussed interleaving and thought of some ways to teach a first-year criminal course that way -- but I was teaching criminal law for the first time and didn't want to assume risk of a full-on implosion my first time through.

Rather than teaching each discreet topic in a linear fashion (everything you need to know about this element, then this crime, then this defense), I thought about teaching with a fact pattern, where we would work through the case and touch on an aspect of each basic element, look at how the elements came together to form a crime, and then maybe look at one defense. Once that was over, we would start all over with a new case and go over more complex aspects of each element, a new defense, etc. Pulling that off takes serious work (I don't know of any casebooks organized that way), and students who are receptive to new things.

I ended up using a hybrid model, where I pulled some topics out of the normal linear, discreet presentation (standards of proof, standards of review, all crimes but homicide), and interleaved that stuff as I went through the various elements in detail. I also ended up presenting homicide as a return to mens rea, rather than as "hey, let's look at this crime." That was about as much risk as I could handle, and the class went pretty well.

If you can do it in con law and civ pro, maybe there is hope for the rest of us!

Posted by: Eric Carpenter | Jul 3, 2015 3:55:27 PM

Post a comment